Abstract
Pulmonary infection is common in connective tissue diseases (CTDs) patients because of immunodeficiency. The basic characteristics of pathogens in this set of patients may differs from immunocompetent patients and largely unclear. We aimed to understand these characteristics by metagenomic next-generation sequencing (mNGS) detection in bronchoalveolar lavage fluid (BALF) from CTDs and explored the primary disease features of this group of patients. Eighty-one CTD patients who were suspected pulmonary infection and received mNGS of BALF as well as conventional microbiologic testing (CMT) were enrolled consecutively. We analysed the types of CTDs, whether accompanied with interstitial lung disease, comparison between performance of mNGS and CMT, and distribution of clinically relevant pathogens, etc. Of the 81 cases, 62 were clinically diagnosed with pulmonary infection. Among all patients, idiopathic inflammatory myopathy accounted for the highest proportion of cases infected, especially anti-MDA5 dermatomyositis and anti-synthetase syndrome. Patients in the pulmonary infection group had been previously treated with higher percentages of anti-rheumatic drugs than those in the non-infection group. The sensitivity of mNGS was higher than that of CMT (80.6% vs. 66.1%). Among the microbes detected by mNGS, the most common bacterial pathogen was Pseudomonas aeruginosa, and the most frequently fungi was Pneumocystis jirovecii. As for the specific pathogens, mNGS had great advantages over CMT in identifying Pneumocystis jirovecii. Idiopathic inflammatory myopathy was the disease most susceptible to pulmonary infections among CTDs. mNGS showed high efficiency for the detection of pathogens that cause pneumonia in BALF from patients with CTDs, especially for Pneumocystis jirovecii.